WiSER introduces a transmitter-conditioned sparse 3D scene encoder queried by a ray-corridor decoder for radiomaps and a DETR-style set decoder for variable-cardinality CIR taps, trained on co-registered ScanNet++ and Sionna data.
hub
Large wireless model (LWM): A foundation model for wireless channels
15 Pith papers cite this work. Polarity classification is still indexing.
15
Pith papers citing it
hub tools
representative citing papers
ConsisFormer reduces WFM Transformer complexity by over 83% via adaptive token aggregation and feature interpolation while preserving performance on channel tasks.
LWM-CDE creates a structured representation space for wireless datasets using a foundation model that correlates better with empirical transfer performance than prior metrics.
SPA-MAE adapts an MAE backbone with a physical prior module providing parameter-aware and structure-aware guidance to pretrain on CSI data, yielding better downstream performance than prior CSI foundation models with fewer parameters.
SiFo pretrains a CSI feedback model on source sites and uses RSRP-based user matching to calibration memory for site-specific subspace guidance at target sites without parameter updates.
Channel intrinsic dimensionality dNL (5-35) sets the scaling ceiling for wireless foundation models, with diminishing returns past ~30M parameters and pilot-aided test-time training on 12M models beating 96M static models by 7-10 dB.
RFPrompt adapts the Large Wireless Model via deep prompt tokens to improve out-of-distribution robustness in modulation classification while training only a small number of parameters.
A pre-trained interference-aware graph Transformer model for wireless resource allocation that achieves strong few-shot adaptation to new tasks and scenarios.
A pretrained wireless foundation model claims to unify zero-shot channel reconstruction and few-shot adaptation across 9 CSI tasks, outperforming task-specific supervised baselines.
A two-stage reinforcement learning system on pretrained LLMs aligns channel state information with user intents to generate adaptive, physically realizable link construction strategies for 6G that outperform conventional methods in experiments.
A unified framework for CSI-native foundation models incorporates scale-aware exposure, physical coordinates, and correlation-bounded attention, reporting over 4 dB NMSE gains in zero-shot tasks and 36.6% spectral efficiency improvement with 7% pilot overhead.
Argues that wireless data's configuration dependence and lack of self-containment make monolithic foundation models unsuitable for AI-native 6G, favoring instead composable agentic architectures.
Enwar 3.0 is an LLM-orchestrated framework that uses a sensor degradation classifier and context-aware agent coordination to achieve over 88% beam selection accuracy, 98% blockage F1-score, and 87% reasoning correctness in mmWave vehicular networks.
Surveys adaptation of foundation models to wireless tasks across off-the-shelf, wireless-native, and agentic paradigms for 6G PHY intelligence and network autonomy.
citing papers explorer
-
WiSER: A Wireless Scene Encoder for Geometry-Grounded Multi-View Wireless Prediction
WiSER introduces a transmitter-conditioned sparse 3D scene encoder queried by a ray-corridor decoder for radiomaps and a DETR-style set decoder for variable-cardinality CIR taps, trained on co-registered ScanNet++ and Sionna data.
-
ConsisFormer: Compute-Efficient Transformer for Wireless Foundation Models Based on Channel Consistency
ConsisFormer reduces WFM Transformer complexity by over 83% via adaptive token aggregation and feature interpolation while preserving performance on channel tasks.
-
LWM-CDE: A Representation Space for Wireless Data Reasoning and Transferability
LWM-CDE creates a structured representation space for wireless datasets using a foundation model that correlates better with empirical transfer performance than prior metrics.
-
SPA-MAE: A Physics-Guided CSI Foundation Model for Wireless Physical Layer
SPA-MAE adapts an MAE backbone with a physical prior module providing parameter-aware and structure-aware guidance to pretrain on CSI data, yielding better downstream performance than prior CSI foundation models with fewer parameters.
-
SiFo: Wireless Foundation Model for Low-Overhead Site-Specific CSI Feedback
SiFo pretrains a CSI feedback model on source sites and uses RSRP-based user matching to calibration memory for site-specific subspace guidance at target sites without parameter updates.
-
How Big Should a Wireless Foundation Model Be?
Channel intrinsic dimensionality dNL (5-35) sets the scaling ceiling for wireless foundation models, with diminishing returns past ~30M parameters and pilot-aided test-time training on 12M models beating 96M static models by 7-10 dB.
-
RFPrompt: Prompt-Based Expert Adaptation of the Large Wireless Model for Modulation Classification
RFPrompt adapts the Large Wireless Model via deep prompt tokens to improve out-of-distribution robustness in modulation classification while training only a small number of parameters.
-
A Graph Foundation Model for Wireless Resource Allocation
A pre-trained interference-aware graph Transformer model for wireless resource allocation that achieves strong few-shot adaptation to new tasks and scenarios.
-
WiFo-2: a generalist foundation model unifies heterogeneous wireless system design
A pretrained wireless foundation model claims to unify zero-shot channel reconstruction and few-shot adaptation across 9 CSI tasks, outperforming task-specific supervised baselines.
-
Agentic Link Construction for Environment and Intent Aware 6G Communication
A two-stage reinforcement learning system on pretrained LLMs aligns channel state information with user intents to generate adaptive, physically realizable link construction strategies for 6G that outperform conventional methods in experiments.
-
Towards CSI-Native Foundation Models: A Channel-Adaptive Roadmap for 6G
A unified framework for CSI-native foundation models incorporates scale-aware exposure, physical coordinates, and correlation-bounded attention, reporting over 4 dB NMSE gains in zero-shot tasks and 36.6% spectral efficiency improvement with 7% pilot overhead.
-
Against the Monolithic Wireless World Model: Why NextG Needs Composable and Agentic Intelligence
Argues that wireless data's configuration dependence and lack of self-containment make monolithic foundation models unsuitable for AI-native 6G, favoring instead composable agentic architectures.
-
Enwar 3.0: An Agentic Multi-Modal LLM Orchestrator for Situation-Aware Beamforming, Blockage Prediction, and Handover Management
Enwar 3.0 is an LLM-orchestrated framework that uses a sensor degradation classifier and context-aware agent coordination to achieve over 88% beam selection accuracy, 98% blockage F1-score, and 87% reasoning correctness in mmWave vehicular networks.
-
Foundation Models for Wireless Communications: From PHY Intelligence to Network Autonomy
Surveys adaptation of foundation models to wireless tasks across off-the-shelf, wireless-native, and agentic paradigms for 6G PHY intelligence and network autonomy.
- WiFo-MiSAC: A Wireless Foundation Model for Multimodal Sensing and Communication Integration via Synesthesia of Machines (SoM)